camber in steel beam 3

[Lutfi]

I have a beam that will span about 60 feet (simple span). Beam DL+LL deflections are better that L/360. I have been thinking about cambering the beam. I have two questions:

1. Should I request and specify camber or should I love with mill camber for this length of span?
2. If I to specify a camber, how much? Just enough for the dead load camber?

Any input and thoughts would be appreciated.


Regards,
Lutfi

 

[haynewp]

I always camber for the dead load I am sure will be there and show it on the plan beside the beam (1 1/2" etc.), I don't think I have ever cambered for live load but some people might. I normally like to camber when it amounts to > 1/2", I don't think I have ever cambered for 1/2" or less. I try not to camber for beams with moment connections at the ends, braced frames, or spandrel beams.


When things are steep, remember to stay level-headed.
-Horace

 

[dik]

If camber required is less than 1" I usually use mill camber. I usually camber for DL or thereabouts. I just recently completed a 60'+ span and cambered it for 2". Picking up part of an existing steel pre-eng bldg frame... column removed.

Dik

 

[SlideRuleEra]

Lutfi - I have considered cambering for both DL + LL, but have not actually done it. This was for elevated industrial equipment slab floor beams. The slab was exposed to the weather. In hind site, wish that I had used camber. With it we would have had positive drainage, instead of ponding that now occurs.

AISC Manual, 9th Edition, has a good discussion about cambering on page 1-147.

Also, see this article in the July 2006, issue of Modern Steel Construction

http://www.aisc.org/MSCTemplate.cfm...Management/ContentDisplay.cfm&ContentID=32224

...and this one in the September 2004 issue (a reprint from 1989)

http://www.aisc.org/MSCTemplate.cfm...Management/ContentDisplay.cfm&ContentID=26794

http://www.SlideRuleEra.net

 

[haynewp]

Here is a pretty good AISC presentation I found;

http://www.aisc.org/Content/ContentGroups/Documents/University_Relations3/CompConstCamber1.ppt

When things are steep, remember to stay level-headed.
-Horace

 

[Lutfi]

Many thanks for the good and quick input.

Regards,
Lutfi

 

[UcfSE]

Take a look at this article in Modern Steel Construction.

MSC Specifying Camber Article

 

[Dinosaur]

I would absolutely camber for a 60 foot span. I would only camber for selfweight + superimposed DL. I would camber regardless of the presence or absence of moment connections. Good Luck.

 

[WillisV]

Referring to Dinosaur's comment - I do not think it is a good idea at all to camber in the presence of moment connections, especially if they are of the full-pen welded variety. You are essentially counting on a connection that is supposed to be perfectly square to not be perfectly square at the same time.

 

[Dinosaur]

OK, I'll bite. Why not?

 

[WillisV]

Fit up - if a beam is cambered each end will be rotated slightly from vertical and as load is applied they will rotate back to vertical - this behaviour cannot be realized in a fully welded moment connection as the connection needs to be a perfectly vertical 90 degree joint to begin with, or if you actually detail the end of the vertical with a slight cant then the welds would prevent the required rotation for the camber to be relieved.

 

[Dinosaur]

In my industry we account for both the fit-up problem and the end rotations effect on the other parts. In bridge design, the spans are of such a length that you have to account for camber.

 

[WillisV]

I realize its possible however for buildings it is just not worth it. AISC also generally recommends against it (see MSC article posted by UcfSE).

 

[pba]

I've specified camber a few times but more commonly add a note leaving the details of how much to the fabricator. Which raises the question - How exactly does the fabricator get the beam to be curved? We are talking about a very large radius. If the beam is simpy bent at the centre does this actually help the deflected shape or make it worse?

 

[WillisV]

pba - read the MSC article posted by UcfSE above.

 

[lkjh345]

A couple of thoughts:

We have done several office type buildings in our office with floor bay spans of 60 to 64'. We camber for 75% of the dead load, and this seems to work out well. Even though the floor beams are modeled as simple spans, the end connections (to girders or columns) do in fact have some rotational restraint. They are not perfect pin connections. The 75% is an fairly arbitrary number, but it seems to have worked well for us over the years.

In addition, we generally factor in an extra 1/2" of concrete in our loads to allow the contractor to pour a level floor. We specify a 'level floor', not a 'constant thickness' floor. Again, this seems to have worked out well for us over the years.

 

[craigory28]

There is a lot of good advice listed out here already, but I have one tidbit I would like to add. I do not camber for perimeter beams. Especially ones that have an angle for wall attachment or pour stop. The reason is even though we don't design for the angle to help us from a stress point. It is there for a stiffness point. (It actually greatly increases your I. You will not get your camber out of the beam.